200401348 玖、發明說朋” 【發明所屬之技術領域】 本發明係關於光罩及半導體基板等的圖案描畫方法，尤 其是關於使用可變成形向量掃描方式的電子束描書裝胃白勺 圖案描畫方法。 【先前技術】 圖12(A)〜圖12(C)爲說明使用可變成形向量掃描方式 的電子束描畫裝置的習知圖案描畫方法用的模式圖。在欲 描畫圖1 2 (A)所示圖案F的情況，首先，藉由第1次的照 像拍攝描畫長方形的區域F1 (圖12(B))，其次，藉由第2 次的照像拍攝描畫鄰接區域F1的長方形的區域F2(圖 12(C))。藉此，作爲2個長方形圖案的組合，而獲得圖12(A) 所示圖案F。尙，描畫區域F 1用的電子束的照射時間（以 下稱爲「照像拍攝時間」）及描畫區域F2用的照像拍攝時 間相等。 【發明內容】 (發明所欲解決之問題） 但是，如此在藉由多次照像拍攝進行1個圖案的描畫的 情況，相對於圖案的設計尺寸而有於精加工尺寸上產生誤 差的問題。尤其是，如圖1 2(c)所示，在伴隨著描畫相關 於Y方向的照像拍攝尺寸W小的區域F 2的步驟的情況’ 確認到圖案F的精加工尺寸較設計尺寸極爲增大。這可以 認爲是所照射的電子束的光束輪廓在區域F 1及區域F2相 互差異，及每一照像拍攝尺寸的電流密度具有差異等的原 326\專利說明書(補件)\92-06\92105345 200401348 因。尙，圖12中，規定紙面的上下方向及左右方向分別爲 「X方向」及「γ方向」。 但是，作爲校正起因於照像拍攝尺寸的精加工尺寸的誤 差的方法’眾知有照像拍攝等級校正的方法。照像拍攝等 級校正係爲按照照像拍攝尺寸而使照像拍攝時間（即，放射 厘）變化進行描畫，藉以校正誤差的方法。圖1 3 ( A )及圖 1 3(B)爲說明照像拍攝等級校正用的圖。參照圖i3(a)，假 定孤立圖案的描畫，將相關於Y方向的電子束的尺寸定義 爲照像拍攝尺寸S。圖1 3 (B)的曲線圖的橫軸爲照像拍攝尺 寸S，縱軸爲誤差G(即，相關於γ方向，從圖案的精加工 尺勺減去設計尺寸的値）。參照圖1 3 ( B )，在照像拍攝尺寸 s小的特定的範圍，可知精加工尺寸較設計尺寸變小。這 是受到近接效應的影響。據此，在照像拍攝尺寸s爲例如 S 2的情況以標準的放射量（特性q〗）進行描畫時，產生—G 1 的誤差。該情況’藉由利用較標準高的放射量（特性q 2)以 進行描畫的方式來校正，即可消除誤差-G 1。尙，每一照 像拍攝尺寸S的放射量的校正量，藉由實驗可求取。 如此之照像拍攝等級校正，係爲在圖案的精加工尺寸較 設計尺寸小的情況，藉由較標準高的放射量進行描畫，校 正精加工尺寸的誤差者。相對於此，在如圖1 2所示習知的 圖案描畫方法中，如上述，圖案F的精加工尺寸較設計尺 寸大。據此，無法藉由照像拍攝等級校正來校正根據習知 的圖案描畫方法的精加工尺寸的誤差。 本發明係爲了解決習知的圖案描畫方法中的上述問題 ;32(5\專利說明書(補件)\92-〇6\92105345 200401348 1'占而元成者，其目的在於’在藉由多次照像拍攝來描書一 個圖案的情況，獲得可適宜校正精加工尺寸的相對於設計 尺寸的誤差的圖案描靈方法。 (解決問題之手段） 本發明中’申請專利範圍第1項記載之圖案描畫方法， 具備如下的步驟，（a)藉由將電子束第！時間照射於描畫對 象的第1區域，以描畫具有沿第1方向延伸的第1邊及沿 第2方向延伸的第2邊的第1圖案的步驟；（b)作爲與步驟 (a)不同的步驟予以執行’藉由於描畫對象的第2區域照射 較第1時間短的第2時間電子束，以描畫具有重疊於第1 圖案之第1邊的第1邊及沿第2方向延伸的第2邊，且該 第2邊的尺寸在指定的臨限値以下的第2圖案的步驟。 此外’本發明中’申請專利範圍第2項記載之圖案描畫 方法’係於申請專利範圍第1項記載之圖案描畫方法，其 特徵爲：臨限値爲0.3 /z m。 【實施方式】 圖1爲顯示本發明之實施形態之圖案描畫裝置的整體結 構的槪略方塊圖。本實施形態之圖案描畫裝置具備可變成 形向量掃描方式的電子束描畫裝置1、資料處理部2及記 憶部3。 圖2(A)、圖2(B)〜圖4爲說明被記憶於記憶部3的資料 用的圖。如上述，在藉由多次照像拍攝進行〗個圖案的描 畫的情況’根據照像拍攝尺寸，圖案的精加工尺寸較設計 尺寸極爲增大。在此’首先’劃分出影響波及精加工尺寸 326\專利說明書(補件)\92-〇6\92105345 200401348 的照像拍攝尺寸的範圍。圖2 (A)顯示相關於γ方向的設計 尺寸爲一定値的圖案Κ。在圖2(Α)的例子中，相關於γ方 向的設計尺寸等於最大照像拍攝尺寸（例如爲2 〇 # m)。藉 由2次照像拍攝描畫該圖案κ ^具體而言，藉由第1次照 像拍攝描畫圖案κ 1，藉由第2次照像拍攝插畫圖案κ 2。 然後，使照像拍攝尺寸S作各式各樣的變化，於每一照像 拍攝尺寸S測定相關於γ方向的圖案κ的精加工尺寸。在 此，照像拍攝尺寸S係規定爲第2次照像拍攝的相關於γ 方向的尺、j ’等於圖案Κ 2的設計尺寸。此外，描書圖案 Κ 1、K2時的照像拍攝時間τ，均爲標準的照像拍攝時間 Τ卜 其次’基於上述實驗的結果，製成圖2(Β)所示曲線圖。 上述橫軸爲照像拍攝尺寸S ( # m)，縱軸爲誤差G ( # m)。 誤差G係爲相關於γ方向，作爲從圖案κ的精加工尺寸減 去設計尺寸（本例爲2.0 a m)的値所獲得。參照圖2(B)，在 照像拍攝尺寸S爲0.3 m以下的範圍，可知照像拍攝尺 寸S越小則誤差g越爲增大。也就是說，瞭解到影響波及 精加工尺寸的照像拍攝尺寸s範圍，在〇 · 3 # m以下。其 結果’將校正所必須的照像拍攝尺寸的臨限値S 0設定在 0.3 μ m。 再者’藉由階段性縮短描畫圖案K2時的照像拍攝時間T 進行與上述相同的實驗，於每一照像拍攝尺寸S求得取消 誤差G所獲的照像拍攝時間的校正量。在此，描畫圖案 K 1時的照像拍攝時間τ，就爲標準的照像拍攝時間T 1。 9 326\專利說明書(補件)\92-〇6\92105345 200401348 寥照圖3，特性u 1係爲以標準照像拍攝時間Τ ]照射電子 束的情況獲得的特性，特性U2、U 3係爲以較標準照像拍 攝時間Τ 1短的照像拍攝時間Τ2、Τ3照射電子束的情況所 獲得的特性。根據圖3，可知在照像拍攝尺寸S爲S 2 ( // m)的情況’藉由以對應於特性U 2的照像拍攝時間Τ 2照射 電子束’可以消除誤差G1。此外’可知在照像拍攝尺寸S 爲S3(以m)的情況，藉由以照像拍攝時間T3照射電子束， 可以消除誤差G2。 再者’根據圖3所示曲線圖，製成圖4所示校正表格4。 校正表格4記述照像拍攝尺寸S及在採用該照像拍攝尺寸 S的情況取消誤差G所獲得的照像拍攝時間T的對應關 係。圖1所示記憶部3記憶著上述校正表格4相關的資料。 圖5及圖6(A)〜圖6(D)爲說明本實施形態之圖案描畫方 法用的流程圖及模式圖。以下，參照圖1、5、6(A)〜圖 6(D) ’按步驟順序說明本實施形態之圖案描畫方法。尙， 圖4所示校正表格4已被記憶在記憶部3。 首先，在步驟SP1，資料處理部2係從外部輸入欲描畫 之圖案（以下稱爲「描畫圖案」）的形狀資料。圖6 ( A )顯示 描畫圖案P的一例。在步驟SP 1，將上述照像拍攝尺寸之 臨限値S 0相關之資料及描畫圖案之位置相關之資料一倂 輸入資料處理部2。此外，除此等資料外，也可輸入標準 照像拍攝時間T 1相關的資料。 其次’在步驟SP2，資料處理部2係基於描畫圖案之資 料來決定描畫的照像拍攝區域。在此，「照像拍攝區域」係 326\專利說明書(補件)\92-06\921〇5345 200401348 意味著以一次照像拍攝所描畫的區域。此時，即使在最大 照像拍攝尺寸以下也無法以一次照像拍攝描畫的描畫圖 案’係被分割爲連續的多個照像拍攝區域。參照圖6 (B )， 描畫圖案P被分割爲連續的照像拍攝區域Η丨、H2。照像 拍攝區域Η 1、Η 2係以互異的照像拍攝進行描畫。 再者’在步驟S Ρ 3，資料處理部2係從多個照像拍攝區 域中抽出照像拍攝時間的校正所必要的照像拍攝區域。在 此’「校正」係意味著在起因於將一個圖案分割爲連續的多 個照像拍攝區域而使得圖案的精加工尺寸變爲較設計尺寸 大的情況，用以縮短照像拍攝時間的校正。以下，稱如此 之校正爲「起因於照像拍攝分割的校正」。資料處理部2 係基於照像拍攝尺寸之臨限値S 0 ’抽出γ方向相關之照像 拍攝尺寸爲0.3 # m以下的照像拍攝區域。因爲照像拍攝 區域Η 2的照像拍攝尺寸S 2較0.3 # m還要小，因此，如 圖6 ( C )所示’照像拍攝區域Η 2係作爲起因於照像拍攝分 割的校正所必要的照像拍攝區域而被抽出。 具體而言，資料處理部2係將同時滿足照像拍攝區域的 任意一邊的照像拍攝尺寸在臨限値S0以下（條件R1)以及 上述任意的一邊和與其對向的邊以外的邊繫接於其他的照 像拍攝區域（條件R2)的2個條件的照像拍攝區域，作爲起 因於照像拍攝分割的校正所必要的照像拍攝區域而予以抽 出。 圖7〜圖9爲顯示照像拍攝區域的抽出例的模式圖。圖 7係對應孤立圖案而顯示關於Y方向之照像拍攝尺寸爲0.3 11 326\專利說明書(補件)\92-〇6\92105345 200401348 // ιή以下的照像拍攝區域η 3。圖8顯示X方向及Y方向 相關之照像拍攝尺寸均較0.3 ν m大的照像拍攝區域Η 4， 及和照像拍攝區域Η4分開’且γ方向相關之照像拍攝尺 寸在0 · 3 μ m以下的照像拍攝區域η 5。圖9顯示X方向及 Υ方向相關之照像拍攝尺寸均較〇.3 μ ηι大的照像拍攝區 域Η 6 ’及和照像拍攝區域H 6連續，且X方向相關之照像 拍攝尺寸在〇 . 3 v m以下的照像拍攝區域η 7。因爲照像拍 攝區域Η 4、Η 6未滿足上述條件ri，因而未被作爲起因於 照像拍攝分割的校正所必要的照像拍攝區域抽出。此外， 因爲照像拍攝區域Η3、Η5、Η7未滿足上述條件R2 ,因而 未被作爲起因於照像拍攝分割的校正所必要的照像拍攝區 域抽出。 接續步驟SP3 ’在步驟SP4，資料處理部2分別決定各 照像拍攝區域的照像拍攝時間。具體而言，資料處理部2 係藉由參照被記憶於記憶部3的校正表格4，使與各照像 拍攝區域的照像拍攝尺寸s對應，來分別決定照像拍攝區 域T。如圖6(D)所示，將照像拍攝尺寸爲s 1的照像拍攝 區域Η 1的照像拍攝時間設定爲標準的照像拍攝時間τ !， 將照像拍攝尺寸爲S2的照像拍攝區域Η2的照像拍攝時間 設定爲照像拍攝時間Τ2。照像拍攝區域與照像拍攝時間的 關係’是作爲資料而從資料處理部2輸入記憶部3，被記 億於記憶部3。 此外，參照圖1，資料D係從資料處理部2被輸入電子 束描畫裝置1。電子束描畫裝置1係基於輸入的資料D， —326\專利說明書(補件)\92-06\921〇5345 200401348 執行圖案的描畫。圖1 0爲顯示藉由電子束描畫裝置1執行 圖案描畫的狀況的立體圖。光罩基板等的描畫對象1 〇係載 置於未圖示的載台上，描畫對象1 〇上照射著電子束1 1。 圖1 1 (A)及圖1 1 (Β )與圖6對應’爲顯示於描畫對象i 〇 上描畫圖案的步驟的模式圖。參照圖1 1 (A)，首先，在對 應於圖6(B)所示照像拍攝區域H1的描畫對象10內的區域 (以下稱爲「第1區域」），以標準的照像拍攝時間T1照射 電子束]1。藉此，在描畫對象1 〇的第1區域描畫圖案Μ ]。 圖案Μ 1具有沿著X方向延伸的邊L1及沿著Υ方向延伸 的邊L2。 再者，參照圖11(B)，在對應於圖6(C)所示照像拍攝區 域Η2的描畫對象1 0內的區域（以下稱爲「第2區域」）， 以照像拍攝時間Τ2照射電子束1 1。在此，第2區域被規 定爲鄰接圖案Ml的邊L1的區域。藉此，在描畫對象10 的第2區域描畫圖案M2。圖案M2具有沿著X方向延伸的 邊L3及沿著Y方向延伸的邊L4。圖案M2的邊L3係重疊 於圖案Μ 1的邊L1。藉由校正照像拍攝時間的效果，使得 Υ方向相關之圖案Μ 2的精加工尺寸與設計尺寸（亦即，Υ 方向相關之照像拍攝區域Η2的照像拍攝尺寸S2)成爲相 等。 尙，在以上的說明中，針對描畫圖案Μ 1後再描畫圖案 Μ 2的情況的例子做了敘述，但是，相反地，也可在描畫 圖案M2後再描畫圖案Ml。 在半導體裝置之製程之一的光微影步驟中，使用藉由電 13 3如\專利說明書(補件)\92-06\92而345 200401348 子束描畫裝置1所製作的光罩，使用如KrF激勵雷射光作 爲曝露光，對於形成於半導體基板上的光阻進行縮小投影 曝光。但是，也可不使用光罩而藉由電子束直接將圖案描 畫於半導體基板上。該情況，圖1 〇所示描畫對象1 0形成 半導體基板。 根據如此之本實施形態之圖案描畫方法，在將一個圖案 分割爲連續的多個照像拍攝區域，藉由多次照像拍攝以描 畫一個圖案的情況，關於照像拍攝尺寸在指定的臨限値以 下的照像拍攝區域，按照其照像拍攝尺寸縮短校正照像拍 攝時間。藉此，可減低精加工尺寸對於圖案的設計尺寸的 誤差。 尙，圖4所示校正表格4最好按照光阻的種類及光罩的 遮光膜的種類等準備多個。該情況，在如圖5所示流程的 步驟SP4中，資料處理部2從多個校正表格4中選擇按照 所採用的光阻的種類等的合適的校正表格4。 此外，資料處理部2也可不僅考慮起因於照像拍攝分割 的校正，還考慮近接效應校正及糢糊化（Fogging)校正等的 其他的校正，來決定各照像拍攝區域的照像拍攝時間。 又’關於未成爲起因於照像拍攝分割的校正的對象的照 像拍攝區域中需要照像拍攝等級校正者（例如，圖7〜圖9 所示照像拍攝區域Η 3、Η 5、Η 7 )，也可進行照像拍攝等級 校正。 (發明效果） 根據本發明之申請專利範圍第1、2項之發明，在描畫 326\專利說明書(補件)\92-06\92105345 200401348 第2邊的尺寸爲指定的臨限値以下的第2圖案時，藉由縮 短電子束的照射時間，即可減低精加工尺寸對於圖案的設 計尺寸白勺誤差。 【圖式簡單說明] 圖1爲顯示本發明之實施形態之圖案描畫裝置的整體結 構的槪略方塊圖。 圖2(A)及圖2(B)爲說明被記億於記憶部的資料用的圖。 圖3爲說明被記憶於記憶部的資料用的圖。 圖4爲說明被記憶於記憶部的資料用的圖。 圖5爲說明本發明之實施形態之圖案描畫方法用的流程 圖。 圖6(A)〜圖6(D)爲說明本發明之實施形態之圖案描畫 方法用的模式圖。 圖7爲顯不照像拍攝區域的抽出例的模式圖。 圖8爲顯示照像拍攝區域的抽出例的模式圖。 圖9爲顯示照像拍攝區域的抽出例的模式圖。 圖10爲顯示藉由電子束描畫裝置執行圖案描畫的狀況 的立體圖。 圖11(A)及圖11(B)爲顯示於描畫對象描畫圖案的步驟 的模式圖。 圖12(A)〜圖12(C)爲說明習知圖案描畫方法用的模式 圖。 圖13(A)及圖13(B)爲說明照像拍攝等級校正用的圖。 (元件符號說明） 326\專利說明書(補件)\92-06\921 〇5345 ί~ν υ匕丨ο 〇 200401348 1 電子束描畫裝置 2 資料處理部 3 記憶部 4 校正表格 I 〇 描畫對象 II 電子束 D 資料 G 誤差 G1 誤差 G2 誤差 H1 照像拍攝區域 Η 2 照像拍攝區域 Η 3 照像拍攝區域 Η4 照像拍攝區域 Η 5 照像拍攝區域 Η 6 照像拍攝區域 Η 7 照像拍攝區域 Κ 圖案 Κ1 第1次照像拍攝描畫圖案 Κ2 第2次照像拍攝描畫圖案 L1 沿著X方向延伸的邊 L2 沿著Υ方向延伸的邊 L3 沿著X方向延伸的邊 L4 沿著Υ方向延伸的邊 326\專利說明書(補件)\92-06\92105345 200401348200401348 "The invention belongs to friends" [Technical field to which the invention belongs] The present invention relates to a pattern drawing method for a photomask, a semiconductor substrate, and the like, and more particularly, to a pattern drawing of a stomach mounted on an electron beam using a variable forming vector scanning method Method [Prior art] Fig. 12 (A) to Fig. 12 (C) are schematic diagrams for explaining a conventional pattern drawing method using an electron beam drawing device of a variable forming vector scanning method. In the case of the pattern F shown in the figure, first, the rectangular region F1 is drawn by the first photographing (FIG. 12 (B)), and second, the rectangular region of the adjacent region F1 is drawn by the second photographing. Area F2 (FIG. 12 (C)). As a result, the pattern F shown in FIG. 12 (A) is obtained as a combination of two rectangular patterns. Alas, the irradiation time of the electron beam for drawing the area F1 (hereinafter referred to as "Photographing time") and the photographic time for drawing area F2 is equal. [Summary of the Invention] (Problems to be Solved by the Invention) However, in the case where a pattern is drawn by taking multiple photographs as described above, there is a problem that an error occurs in the finishing size with respect to the design size of the pattern. In particular, as shown in FIG. 12 (c), in the case of the step of drawing a region F 2 with a small photographing size W in the Y direction, it was confirmed that the finishing size of the pattern F is extremely increased compared to the design size. Big. This can be considered as the original 326 \ Patent Specification (Supplement) \ 92-06 where the beam profiles of the irradiated electron beams are different from each other in the area F1 and the area F2, and the current density of each photographed size is different. \ 92105345 200401348. That is, in Fig. 12, the up-down direction and the left-right direction of the paper surface are defined as "X direction" and "γ direction", respectively. However, as a method of correcting an error in a finished size due to a photographing size, a method of correcting a photographing level is known. The photographic level correction is a method of correcting the error by drawing and changing the photographing time (that is, the radiometer) according to the photographing size. Fig. 13 (A) and Fig. 13 (B) are diagrams for explaining correction of a photographing level. Referring to FIG. I3 (a), assuming the drawing of an isolated pattern, the size of the electron beam related to the Y direction is defined as the photographing size S. The horizontal axis of the graph in FIG. 13 (B) is the photographing size S, and the vertical axis is the error G (that is, in relation to the γ direction, the design dimension 减去 is subtracted from the finished ruler of the pattern). Referring to FIG. 13 (B), in a specific range where the photographing size s is small, it can be seen that the finishing size becomes smaller than the design size. This is affected by the proximity effect. According to this, when the photographing size s is, for example, S 2, the standard radiation dose (characteristic q) is used for drawing, an error of −G 1 occurs. In this case, the error -G 1 can be eliminated by correcting the radiation amount (characteristic q 2) with a higher standard than by drawing. Alas, the correction amount of the radiation amount for each photographing size S can be obtained through experiments. This kind of photo shooting level correction is for the case where the finishing size of the pattern is smaller than the designed size, and the radiation amount is higher than the standard for drawing to correct the error of the finishing size. On the other hand, in the conventional pattern drawing method shown in FIG. 12, as described above, the finishing size of the pattern F is larger than the design size. According to this, it is not possible to correct errors in the finishing size according to the conventional pattern drawing method by correcting the photographing level. The present invention is to solve the above-mentioned problems in the conventional pattern drawing method; 32 (5 \ Patent Specification (Supplements) \ 92-〇6 \ 92105345 200401348 1's accounted for, and its purpose is to 'in In the case of drawing a pattern by taking a second photo, a pattern drawing method capable of appropriately correcting the error of the finished size relative to the design size is obtained. (Means for Solving the Problem) In the present invention, one of the items described in the first patent application range The pattern drawing method includes the following steps: (a) irradiating the first area of the drawing object with the electron beam for the first time to draw a first side extending in the first direction and a second side extending in the second direction Step of the first pattern of the edge; (b) Performed as a step different from step (a) 'The second time region of the drawing object is irradiated with a second time electron beam shorter than the first time to draw the second time 1 The second pattern of the first side of the first side of the pattern and the second side extending along the second direction, and the size of the second side is below the specified threshold. In addition, a patent is applied for 'in the present invention' Pattern drawing method described in item 2 of the scope ' The pattern drawing method described in item 1 of the scope of the patent application is characterized in that the threshold value is 0.3 / zm. [Embodiment] FIG. 1 is a schematic block diagram showing the overall structure of a pattern drawing device according to an embodiment of the present invention. The pattern drawing device of this embodiment includes an electron beam drawing device 1, a data processing section 2 and a storage section 3 of a variable forming vector scanning method. Fig. 2 (A), Fig. 2 (B) to Fig. 4 are descriptions stored in A drawing for the data of the memory section 3. As described above, when drawing multiple patterns through multiple photographic shots, 'the finished size of the pattern is extremely larger than the designed size based on the photographed size. Here' First 'divide out the range of photographing sizes that affect the finishing size 326 \ Patent Specification (Supplements) \ 92-〇6 \ 92105345 200401348. Figure 2 (A) shows that the design size related to the γ direction is a certain 値Pattern K. In the example of FIG. 2 (A), the design size related to the γ direction is equal to the maximum photographic shooting size (for example, 2 〇 # m). The pattern is drawn by two photographings. ^ , With the first photo shoot Draw a pattern κ 1 and shoot an illustration pattern κ 2 with the second photograph. Then, change the photographing size S in various ways, and measure the pattern κ related to the γ direction at each photographing size S. Here, the photographing size S is defined as the ruler in the γ direction and j ′ equal to the design size of the pattern K 2 for the second photographing. In addition, when the tracing patterns K 1 and K 2 The photographing time τ is the standard photographing time Tb. Next, based on the results of the above experiment, the graph shown in Figure 2 (B) is made. The horizontal axis is the photographing size S (# m) , The vertical axis is the error G (# m). The error G is obtained by subtracting the design dimension (2.0 a m in this example) from the finishing dimension of the pattern κ with respect to the γ direction. Referring to Fig. 2 (B), in a range where the photographing size S is 0.3 m or less, it can be seen that the smaller the photographing size S is, the larger the error g is. In other words, it is understood that the range of the photographing size s that affects the finishing size is below 0.3 m. As a result, the threshold 値 S 0 of the photographing size necessary for correction is set to 0.3 μm. Furthermore, the same experiment as described above is performed by stepwise shortening the photographing time T when the pattern K2 is drawn, and the correction amount of the photographing time obtained by canceling the error G is obtained for each photographing size S. Here, the photographing time τ when the pattern K 1 is drawn is the standard photographing time T 1. 9 326 \ Patent Specification (Supplement) \ 92-〇6 \ 92105345 200401348 As shown in Fig. 3, the characteristic u 1 is a characteristic obtained when the electron beam is irradiated at a standard photographing time T], and the characteristics U2 and U 3 are This is a characteristic obtained in a case where an electron beam is irradiated with a photo taking time T2, T3 shorter than the standard photo taking time T1. From Fig. 3, it can be seen that in a case where the photographing size S is S 2 (// m) ', the error G1 can be eliminated by irradiating the electron beam at the photographing time T 2 corresponding to the characteristic U 2. In addition, it can be seen that when the photographing size S is S3 (in m), by irradiating the electron beam at the photographing time T3, the error G2 can be eliminated. Furthermore, according to the graph shown in Fig. 3, a correction table 4 shown in Fig. 4 is prepared. The correction table 4 describes the correspondence between the photographing size S and the photographing time T obtained by canceling the error G when the photographing size S is used. The storage unit 3 shown in FIG. 1 stores data related to the correction table 4 described above. 5 and 6 (A) to 6 (D) are a flowchart and a schematic diagram for explaining a pattern drawing method in this embodiment. Hereinafter, the pattern drawing method according to this embodiment will be described in order of steps with reference to Figs. 1, 5, 6 (A) to 6 (D) '. Alas, the correction table 4 shown in FIG. 4 has been stored in the storage section 3. First, in step SP1, the data processing unit 2 inputs shape data of a pattern to be drawn (hereinafter referred to as "drawing pattern") from the outside. Fig. 6 (A) shows an example of the drawing pattern P. In step SP1, the data related to the above-mentioned photographing size limit 値 S0 and the data related to the position of the drawing pattern are input to the data processing section 2 at once. In addition, in addition to these data, it is also possible to input data related to the standard photo shooting time T 1. Next, at step SP2, the data processing unit 2 determines a photographed shooting area based on the data of the drawing pattern. Here, "photographing area" means 326 \ Patent Specification (Supplement) \ 92-06 \ 921〇5345 200401348, which means that the area drawn by one photographing is taken. At this time, even if the size is smaller than the maximum photographing size, the drawing pattern that cannot be drawn in one photograph is divided into a plurality of consecutive photographing areas. Referring to FIG. 6 (B), the drawing pattern P is divided into continuous photographing regions Η and H2. Photo Shooting areas Η 1, Η 2 are drawn with different photo shoots. Furthermore, in step SP3, the data processing unit 2 extracts a photographing area necessary for correction of photographing time from a plurality of photographing areas. Here, "" correction "means the correction for shortening the photographing time when the finishing size of a pattern becomes larger than the design size due to the division of a pattern into a plurality of continuous photographing regions. . Hereinafter, such a correction will be referred to as "correction due to division of a photograph". The data processing unit 2 is based on the threshold of the photographing size 値 S 0 ′ and extracts photographs related to the γ direction. The photographing area is a photographing area with a size of 0.3 # m or less. Since the photographing size S 2 of the photographing area Η 2 is smaller than 0.3 # m, as shown in FIG. 6 (C), the photographing area Η 2 is used as a correction unit due to the division of the photographing. The necessary photographing area is extracted. To be more specific, the data processing unit 2 connects the photo shooting size that satisfies both sides of the photo shooting area below the threshold 値 S0 (Condition R1), and connects any of the above sides to a side other than the side opposite to it. The two conditional image capturing areas in the other image capturing areas (condition R2) are extracted as the image capturing areas necessary for the correction of the image capturing division. 7 to 9 are schematic diagrams showing examples of extraction of a photographing area. Fig. 7 shows a photographing size corresponding to an isolated pattern in the Y direction of 0.3 11 326 \ Patent Specification (Supplement) \ 92-〇6 \ 92105345 200401348 // photographing area η 3 or less. FIG. 8 shows that the photographing size of the photographs related to the X direction and the Y direction are larger than the photographing area Η 4 which is larger than 0.3 ν m, and separated from the photographing area Η 4 ′, and the photographing size of the photographs related to the γ direction is 0 · 3 Photographing area η 5 or less. FIG. 9 shows that the photographing size related to the X direction and the Υ direction is larger than the photographing area Η 6 ′ and the photographing area H 6 which are continuous, and the photographing size related to the X direction is between 0.3 photographic shooting area n 7 or less. Since the photo shooting areas Η 4, Η 6 do not satisfy the above condition ri, they are not extracted as the photo shooting areas necessary for the correction due to the photo shooting division. In addition, since the photo shooting areas Η3, Η5, and Η7 do not satisfy the above-mentioned condition R2, they are not extracted as the photo shooting areas necessary for the correction due to the photo shooting division. Continuing step SP3 'In step SP4, the data processing unit 2 determines the photographing time of each photographing area. Specifically, the data processing unit 2 refers to the correction table 4 stored in the storage unit 3, and determines the photographing area T by corresponding to the photographing size s of each photographing area. As shown in FIG. 6 (D), the photo shooting time of the photo shooting area Η 1 is set to the standard photo shooting time τ !, and the photo shooting size is S2. The photographing time of the photographing area Η2 is set as the photographing time T2. The relationship between the photographing area and the photographing time 'is input to the memory section 3 from the data processing section 2 as data, and is recorded in the memory section 3 billion. In addition, referring to Fig. 1, the data D is input from the data processing unit 2 into the electron beam drawing device 1. The electron beam drawing device 1 performs drawing of a pattern based on the input data D-326 \ Patent Specification (Supplement) \ 92-06 \ 921〇5345 200401348. FIG. 10 is a perspective view showing a state where pattern drawing is performed by the electron beam drawing device 1. As shown in FIG. A drawing object 10 such as a photomask substrate is placed on a stage (not shown), and the drawing object 10 is irradiated with an electron beam 11. FIG. 11 (A) and FIG. 11 (B) correspond to FIG. 6 'and are schematic diagrams showing the steps of drawing a pattern on the drawing object i0. Referring to FIG. 11 (A), first, in a region (hereinafter, referred to as a “first region”) within the drawing object 10 corresponding to the photographing region H1 shown in FIG. 6 (B), a standard photographing time is used. T1 irradiates the electron beam] 1. Thereby, a pattern M] is drawn in the first area of the drawing object 10]. The pattern M1 has a side L1 extending in the X direction and a side L2 extending in the Υ direction. Further, referring to FIG. 11 (B), the area (hereinafter referred to as “second area”) within the drawing object 10 corresponding to the photographing area Η2 shown in FIG. 6 (C) is taken at the photographing time T2 Irradiate the electron beam 1 1. Here, the second region is defined as a region adjacent to the side L1 of the pattern M1. Thereby, the pattern M2 is drawn in the second area of the drawing object 10. The pattern M2 has a side L3 extending in the X direction and a side L4 extending in the Y direction. The side L3 of the pattern M2 is superimposed on the side L1 of the pattern M1. By correcting the effect of the photographing time, the finishing size and design size of the pattern M 2 related to the Υ direction (that is, the photographing size S 2 of the photographing area Η 2 with respect to the 方向 direction) are made equal. That is, in the above description, an example of the case where the pattern M2 is drawn and then the pattern M2 is described is described. However, on the contrary, the pattern M1 may be drawn after the pattern M2 is drawn. In the photolithography step, which is one of the manufacturing processes of the semiconductor device, a photomask made by using the electricity 13 3 such as \ Patent Specification (Supplement) \ 92-06 \ 92 and 345 200401348 sub-beam drawing device 1 is used, such as KrF excites the laser light as the exposure light, and reduces the projection exposure of the photoresist formed on the semiconductor substrate. However, a pattern may be directly drawn on the semiconductor substrate by an electron beam without using a photomask. In this case, the drawing object 10 shown in FIG. 10 forms a semiconductor substrate. According to the pattern drawing method of this embodiment, when a pattern is divided into a plurality of continuous photographing shooting regions, and a pattern is drawn by multiple photographing, the photographing size is within a specified threshold.缩短 In the following photo shooting area, shorten the time for correcting the photo according to the size of the photo shooting. Thereby, it is possible to reduce the error of the finished size from the design size of the pattern. Alas, it is preferable to prepare a plurality of correction tables 4 shown in FIG. 4 according to the type of the photoresist and the type of the light-shielding film of the photomask. In this case, in step SP4 of the flow shown in FIG. 5, the data processing unit 2 selects an appropriate correction table 4 according to the type of photoresist used, among the plurality of correction tables 4. In addition, the data processing unit 2 may determine not only the correction caused by the photographing segmentation but also other corrections such as proximity effect correction and fogging correction to determine the photographing time of each photographing area. Also, regarding those who do not need to be corrected by the photographing area due to the correction of the photographing segmentation (for example, the photographing area shown in FIGS. 7 to 9 Η 3, Η 5, Η 7 ), You can also correct the shooting level. (Effects of the Invention) According to the inventions in the first and second scopes of the patent application of the present invention, the size of the second side is 326 \ Patent Specification (Supplement) \ 92-06 \ 92105345 200401348. In the case of 2 patterns, by shortening the irradiation time of the electron beam, it is possible to reduce the error in the design dimension of the finished size from the pattern. [Brief description of the drawings] Fig. 1 is a schematic block diagram showing the overall structure of a pattern drawing device according to an embodiment of the present invention. FIG. 2 (A) and FIG. 2 (B) are diagrams for explaining data recorded in the memory section. FIG. 3 is a diagram for explaining data stored in a memory unit. FIG. 4 is a diagram for explaining data stored in a memory section. Fig. 5 is a flowchart for explaining a pattern drawing method according to an embodiment of the present invention. 6 (A) to 6 (D) are schematic diagrams for explaining a pattern drawing method according to an embodiment of the present invention. FIG. 7 is a schematic diagram of an example of extraction of a photographic imaging region. FIG. 8 is a schematic diagram showing an example of extraction of a photographing area. FIG. 9 is a schematic diagram showing an example of extraction of a photographing area. Fig. 10 is a perspective view showing a state where pattern drawing is performed by an electron beam drawing device. Fig. 11 (A) and Fig. 11 (B) are schematic diagrams showing the steps of drawing a pattern on a drawing object. 12 (A) to 12 (C) are schematic diagrams for explaining a conventional pattern drawing method. 13 (A) and 13 (B) are diagrams for explaining correction of a photographing level. (Explanation of component symbols) 326 \ Patent Specification (Supplement) \ 92-06 \ 921 〇5345 ί ~ ν υ⇨ 丨 ο 〇200401348 1 Electron beam drawing device 2 Data processing section 3 Memory section 4 Calibration table I 〇Drawing target II Electron beam D Data G Error G1 Error G2 Error H1 Photographing area Η 2 Photographing area Η 3 Photographing area Η 4 Photographing area Η 5 Photographing area Η 6 Photographing area Η 7 Photographing area Pattern K1 patrol pattern for the first photo shoot kr2 draw pattern for the second photo shoot L1 side L extending in the X direction side L2 extending in the Υ direction side L3 extending in the X direction extending along the Υ direction Edge 326 \ Patent Specification (Supplement) \ 92-06 \ 92105345 200401348
Ml _Α·Α- 弟 1 區 域 -H+ m 素 宣 圖 案 M2 第 2 區 域 -H4* 描 畫 圖 案 S 眧 ^ 1 > % 像 拍 攝 尺 寸 SO 昭 / ) X Ν 像 拍 攝 尺 寸 之 臨 限 SI 昭 /Μ \、 像 拍 攝 尺 寸 S2 昭 y ι \\ 像 拍 攝 尺 寸 T 昭 像 拍 攝 時 間 ΤΙ 標 準 的 眧 像 拍 攝 時 間 Τ2 眧 y 1、、 像 拍 攝 時 間 Τ3 昭 y v \\ 像 拍 攝 時 間 U 1 特 性 U2 特 性 U3 特 性 17 326\專利說明書(補件)\92-06\92105345Ml _Α · Α- Brother 1 area-H + m Su Xuan pattern M2 2nd area-H4 * Draw pattern S 眧 ^ 1 >% Image shooting size SO Zhao /) X Ν Image shooting size limit SI Zhao / Μ \ , Image shooting size S2 y y \\ image shooting size T 像 image shooting time Τ standard standard image shooting time Τ2 眧 y 1, image shooting time Τ3 yyv \\ image shooting time U 1 characteristic U2 characteristic U3 characteristic 17 326 \ Patent Specification (Supplement) \ 92-06 \ 92105345